Hi Mark!

Thanks for putting some stress on the functorial construction code :-)

On Mon, Feb 17, 2014 at 11:12:20PM -0500, Mark Shimozono wrote:
> I am not entirely happy with the way covariant construction functors are 
> applied.
> I think the Cartesian and tensor product constructors should always receive 
> tuples
> of parents/elements/maps

That's what they do, right? The only thing is that, at the level of
the parent/element/morphism classes, the constructors receive the
tuple in the form of a *args.

> and the constructor for QQ-algebra should accept a single parent/element/map.

I haven't checked the details, but this makes sense. AlgebraFunctor
essentially has been unused so far, and we could flesh it up with a
custom __call__ method doing what you suggest, if you think there is a
use case for it.

> It would be a good idea to be able to optionally specify the
> category if it is getting annoying to guess the category.

Indeed.

> It seems very hacky to look for a method with a certain name,
> to tell whether the constructor applies (which is what the code does right 
> now).

Is it any different from ``a*b`` or ``sum([a,b,c])``` which check
(possibly indirectly) whether there is a method called "__add__" and
use it, and bark otherwise?

> What is the list of such construction functors in sage?

You can check the Functorial Constructions section of:

        http://www.sagemath.org/doc/reference/categories/

Among those, the relevant ones are currently: tensor, cartesian
products, and algebras.

Cheers,
                                Nicolas
--
Nicolas M. ThiƩry "Isil" <nthi...@users.sf.net>
http://Nicolas.Thiery.name/

-- 
You received this message because you are subscribed to the Google Groups 
"sage-combinat-devel" group.
To unsubscribe from this group and stop receiving emails from it, send an email 
to sage-combinat-devel+unsubscr...@googlegroups.com.
To post to this group, send email to sage-combinat-devel@googlegroups.com.
Visit this group at http://groups.google.com/group/sage-combinat-devel.
For more options, visit https://groups.google.com/groups/opt_out.

Reply via email to